Practical Nonlinear Model Predictive Control With Kolmogorov–Arnold Network Models
Open Access
- 19 August 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 146745-146756
Abstract
This paper presents a novel approach to practical nonlinear model predictive control (PNMPC) using Kolmogorov–Arnold networks (KANs) as prediction models. KANs are based on the Kolmogorov–Arnold superposition theorem, enabling efficient nonlinear system modeling with fewer parameters compared to traditional neural networks. This characteristic makes KANs particularly suitable for nonlinear model predictive control (NMPC), where computational efficiency is critical. This paper presents a method to efficiently identify nonlinear models from data, proposing a time-delay KAN (TDKAN) architecture, which consists of a KAN associated with tap delays in its input. Furthermore, it proposes the integration of this novel model architecture into a PNMPC algorithm, with the aim of improving control accuracy while reducing computational burden. The method is validated using a pH neutralization process, a benchmark for nonlinear control problems. The results show that the TDKAN-based PNMPC outperforms other PNMPC approaches, such as those that use echo-state networks and long short-term memory networks, providing better setpoint tracking, lower mean squared error, and faster computation times, which demonstrate the potential of KANs for prediction in practical NMPC applications.Keywords
Funding Information
- Coordenação de Aperfeiçoamento de Pessoal de Nível Superior-Brazil (001)
- Brazilian National Council for Scientific and Technological Development (315546/2021-2)
- Brazilian National Agency of Petroleum, Natural Gas and Biofuels (ANP) under the Human Resource Training Program (PRH 2.1)
This publication has 43 references indexed in Scilit:
- KAN-ODEs: Kolmogorov–Arnold network ordinary differential equations for learning dynamical systems and hidden physicsComputer Methods in Applied Mechanics and Engineering, 2024
- Multi-scenario and multi-stage robust NMPC with state estimation application on the Tennessee-Eastman processControl Engineering Practice, 2023
- Physics-informed machine learning for MPC: Application to a batch crystallization processChemical Engineering Research and Design, 2023
- Nonlinear MIMO System Identification with Echo-State NetworksJournal of Control, Automation and Electrical Systems, 2022
- Learning model predictive control with long short‐term memory networksInternational Journal of Robust and Nonlinear Control, 2021
- Deep Koopman model predictive control for enhancing transient stability in power gridsInternational Journal of Robust and Nonlinear Control, 2020
- Model predictive control: Recent developments and future promiseAutomatica, 2014
- Model predictive control: Review of the three decades of developmentInternational Journal of Control, Automation, and Systems, 2011
- Generalized predictive control—Part I. The basic algorithmAutomatica, 1987
- On calculating with B-splinesJournal of Approximation Theory, 1972